Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@chancevision/cli) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

ChanceVision CLI

Analyze images from your terminal with the Chance Vision visual intelligence API.

Use it for screenshot QA, UI understanding, OCR-style extraction, image captioning, and quick visual checks inside shell scripts.

Install

# npm
npm install -g @chancevision/cli

# pnpm
pnpm add -g @chancevision/cli

# yarn
yarn global add @chancevision/cli

# bun
bun add -g @chancevision/cli

# Run without installing
npx @chancevision/cli see --help

Requires Node.js 20 or newer.

Quick Start

Get an API key at chance.vision, then:

export CHANCEVISION_API_KEY="sk-..."

# Describe an image from a URL
chancevision see https://example.com/photo.png

# Analyze a local file
chancevision see ~/Pictures/screenshot.png

# Stream the response in real time
chancevision see -s https://example.com/photo.png

Example output:

The image shows a dashboard with three summary cards, a revenue chart, and a table of recent orders. The main accessibility issue is low contrast in the secondary labels.

Why ChanceVision CLI?

Fast terminal workflow: inspect screenshots, product images, and documents without leaving your shell.
Useful defaults: prints the model answer by default, with --json for raw API responses.
Focused output formats: request markdown or ui_component when you need a specific response shape.
Streaming support: use --stream for real-time output in interactive workflows.
Script-friendly: supports remote URLs, local files, environment-based auth, and raw JSON output.

Authentication

# Option 1: Flag, useful for one-off commands
chancevision see -k "sk-..." https://example.com/img.png

# Option 2: Environment variable, recommended for daily use
export CHANCEVISION_API_KEY="sk-..."
chancevision see https://example.com/img.png

Usage

chancevision see <image> [options]

Sends the image to the model (chance/chance-vision-1.5) for analysis.

Options

Option	Default	Description
`<image>`	(required)	Image URL or local file path to analyze
`-k, --api-key <key>`	—	API key (or use `CHANCEVISION_API_KEY` env var)
`-s, --stream`	`false`	Stream the response as SSE chunks
`--output-format <format>`	—	Output format: `markdown` or `ui_component`
`--json`	`false`	Print the raw JSON API response
`-v, --verbose`	`false`	Show redacted request details and raw chunks

Examples

Basic image analysis

chancevision see https://images.chance.vision/image/revisit.png
chancevision see ~/Pictures/screenshot.png

Streaming

chancevision see -s https://images.chance.vision/image/revisit.png

Raw JSON

chancevision see --json https://example.com/screenshot.png

Output format

chancevision see --output-format markdown https://example.com/screenshot.png
chancevision see --output-format ui_component https://example.com/screenshot.png

Development

git clone https://github.com/chancevision/chancevision-cli
cd chancevision-cli

pnpm install

# Run tests
pnpm test

# Build
pnpm build

# Link globally for local development
pnpm link --global

Before opening a pull request, run:

pnpm lint
pnpm typecheck
pnpm test

See CONTRIBUTING.md for contribution guidelines.

API

This CLI is powered by the Chance Vision API, a visual intelligence platform with streaming support and UI-component output.

POST https://openapi.chance.vision/v1/chat/completions

License

MIT